The study referenced here shows that the year-over-year change in unemployment rate is a strong predictor of whether the incumbent presidential candidate wins the Electoral College vote. Along with the strength of the incumbent’s support at the nominating convention, the year-over-year change in unemployment rate has a better than high probability of predicting the Electoral vote outcome since unemployment data became available prior to the 1932 election.
Given this result and the fact that Senators are voters in the impeachment process, does the year-over-year change in unemployment rate have anything to say about the prospects for Trump’s impeachment? To answer this question, I looked at the year-over-year change in unemployment rate ahead of Nixon’s resignation and Clinton’s impeachment vote in the Senate. The data is plotted in Figure 1.
Figure 1 The Year/Year Change in Unemployment Rate Under Trump is More Like Clinton’s Than Nixon’s
For Nixon, the unemployment data went from being very favorable to very unfavorable over the twelve months ahead of his resignation. The Threshold in the figure is derived in the study referenced above. If the year-over-year change in unemployment rate exceeds 4.3%, then the incumbent candidate loses the Electoral vote all nine times this occurred since 1932. In Nixon’s case, the data approached the threshold early in 1974 and exceeded it three months ahead of his resignation in August. This trajectory roughly coincides with the rise of his disapproval ratings, which plateaued early in 1974 before reaching 65% at the time of his resignation (https://projects.fivethirtyeight.com/trump-approval-ratings/?ex_cid=rrpromo).
For Clinton, the unemployment data remained favorable throughout his impeachment inquiry in the House and trial in the Senate, where he was acquitted. Similarly, his disapproval rating stayed relatively low around 35% during this period.
For Trump, Time has projected (https://time.com/5689285/impeachment-trump-election/) that if he is impeached by the House, his trial in the Senate may conclude in March of 2020. Using this month to anchor the 0th month in the figure, Trump’s data lies much closer to Clinton’s than Nixon’s. His disapproval rating at about 53% currently lies in-between. However, his disapproval rating has been higher than both Nixon’s and Clinton’s throughout his presidency even though the economy has also been favorable the whole time.
Trump’s disapproval rating has been remarkably steady as has the year/year change in unemployment rate. As shown in Figure 2, the year-over-year change in unemployment rate has been below the threshold since 2010. At 115 months as of December 2019, a new record is set every month it remains below the threshold. The last record was set during the Clinton administration. Nixon wasn’t so fortunate. It appears the United State’s ability to extend periods of favorable unemployment rate trends is improving decade-over-decade, and Trump is benefiting from this trend.
Figure 2 History of the Year/Year Change in Unemployment Rate
The unemployment rate is currently 3.5%, and it hasn’t been this low for 50 years. The Threshold will be exceeded if it goes to 4% by the time of the Senate’s vote. Will a new record be set, or will it go above the Threshold and drive Trump from office? There is not much time left for this to happen, and the latest unemployment data suggests Trump will be vindicated by the Senate.